Mycobacterium tuberculosis, the agent of tuberculosis (TB) kills -2 million people annually and is one of the world's leading causes of death. In AIDS patients, the disease exacts an enormous toll, particularly in the developing nations of the world. The rapid global emergence of strains of M. tuberculosis resistant to the major antibiotics used to treat TB poses a serious threat to public health;such strains are potential agents of bioterrorism and have been designated as Category C Priority Pathogens. BCG, a live attenuated strain of Mycobacterium bovis, is the TB vaccine currently used throughout much of the world. This vaccine is not highly efficacious, and it is contraindicated in immunocompromised individuals such as those with AIDS because it can cause disseminated and sometimes fatal disease. Development of a vaccine that is both safe in AIDS patients and more efficacious than BCG is a high priority in the fight against tuberculosis and is the goal of this application. Previous studies from this laboratory demonstrated the first vaccine more potent than BCG in the guinea pig model of pulmonary TB, a stringent model noteworthy for its relevance to human TB. The new vaccine, a recombinant BCG strain overexpressing the M. tuberculosis 30 kDa major secretory protein (rBCG30), is currently in human trials. However, like the BCG strain from which it is derived, this new live vaccine is unsuitable for immunocompromised persons such as AIDS patients. The purpose of this proposal is to develop a modified version of rBCGSO that is safe for immunocompromised individuals yet more potent than BCG. To do this, we are generating growth- regulatable or growth-restricted strains of rBCGSO that are capable of either controlled or self-limited growth in the host. Because the strains can multiply in the host, we anticipate that they will retain the high potency of rBCGSO. Yet the strains will be safe in the immunocompromised host either because their growth can be easily controlled by the host and immediately stopped if necessary merely by withdrawal of an oral factor upon which their growth is dependent or because they have been engineered such that they can not disseminate. We propose to characterize the capacity of these strains to induce protective immunity in the guinea pig model of pulmonary tuberculosis and to compare the level of protective immunity induced by these strains with that induced by the parental BCG strain and the more efficacious rBCGSO. We shall attempt to increase the potency of the new vaccine strains further through a prime-boost regimen that has previously been demonstrated to enhance significantly the immunoprotection conferred by BCG and which should also be completely safe in immunocompromised persons. Finally, we shall confirm the safety of the new recombinant vaccines in an immunocompromised animal host (SCID mouse model).